© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 28922
    Prev
    Next
    INDEX
    Explanations

    the conjunction "and."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    240
    +0.11
    0.5%
    544
    +0.09
    0.4%
    573
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    501
    +0.11
    0.08
    203
    +0.09
    0.07
    240
    +0.09
    0.06
    Negative Logits
    ãģĨ
    -0.57
     âĪ¼
    -0.57
     Unloaded
    -0.55
    /"
    -0.55
     Infantry
    -0.54
     neighb
    -0.54
    recomm
    -0.53
    "},"
    -0.53
     Contribut
    -0.52
     Agency
    -0.52
    POSITIVE LOGITS
    romeda
    1.49
    hra
    1.17
    rea
    1.12
    ré
    1.11
     secondly
    1.05
     guess
    1.04
     THEN
    0.99
    ersen
    0.97
     yet
    0.97
    alus
    0.96
    Activations Density 0.598%

    No Known Activations